Self-complementarity of oligo-2-aminopyridines: A new class of hydrogen-bonded ladders

Article abstract of DOI:10.1021/ja011679p

A new class of hydrogen-bonded ladders based on hydrogen-bonded dimerization of oligo-α-aminopryidines has been demonstrated. Jorgensen's model can be successfully applied to this hydrogen-bonding system in nonpolar solvents. The results show the competitive enthalpy/entropy compensation relationship upon dimerization. Although increasing the number of hydrogen-bonding interactions would enhance the hydrogen-bonding stabilization enthalpy, this stabilization enthalpy per unit would be partially sacrificed to compensate for the entropy loss due to dimerization. These results clearly support the importance of preorganization in designing hydrogen-bonding guest - host molecules.

Full text of DOI:10.1021/ja011679p

Published on Web 03/29/2002
Self-Complementarity of Oligo-2-aminopyridines: A New
Class of Hydrogen-Bonded Ladders
Man-kit Leung,*,† Ashis B. Mandal,† Chih-Chieh Wang,‡ Gene-Hsiang Lee,†
Shie-Ming Peng,*,† Hsing-Ling Cheng,† Guor-Rong Her,† Ito Chao,*,§ Hsiu-Feng Lu,|
Ying-Chieh Sun,| Mei-Ying Shiao, and Pi-Tai Chou*,†
Contribution from the Department of Chemistry, National Taiwan UniVersity,
Taipei 106, Taiwan, Department of Chemistry, Soochow UniVersity, Taipei, Taiwan,
Institute of Chemistry, Academia Sinica, Taipei, Taiwan, Department of Chemistry,
National Taiwan Normal UniVersity, Taipei, Taiwan, and Department of Chemistry,
National Chung Cheng UniVersity, Chia Yi, Taiwan
Received July 10, 2001. Revised Manuscript Received November 15, 2001
Abstract: A new class of hydrogen-bonded ladders based on hydrogen-bonded dimerization of oligo-R-
aminopryidines has been demonstrated. Jorgensen’s model can be successfully applied to this hydrogen-
bonding system in nonpolar solvents. The results show the competitive enthalpy/entropy compensation
relationship upon dimerization. Although increasing the number of hydrogen-bonding interactions would
enhance the hydrogen-bonding stabilization enthalpy, this stabilization enthalpy per unit would be partially
sacrificed to compensate for the entropy loss due to dimerization. These results clearly support the
importance of preorganization in designing hydrogen-bonding guest-host molecules.
artificial ion channels,5 have been reported during the past
decade. In addition, the principle of hydrogen-bonding interac-
tions has also been applied to crystal engineering.6 This requires
self-recognition between identical molecules, a much less com-
mon phenomenon usually restricted to molecules containing
“complementary donor and acceptor units”. Recently, using the
hydrogen-bond array, instead of covalent bonds, to construct
zipperic ladder molecules has been explored.7,8 Polymeric struc-
ture 1 intrigues us because 1 may dimerize to form a belt-shape
zipperic dimer through hydrogen-bonding interactions. Because
1 contains an alternating proton donor/acceptor sequence, ac-
cording to Jorgensen’s theory,9 a relatively strong secondary
repulsive interaction is expected. This would lead to a relatively
small hydrogen-bonding stabilization arising from each of the
repeating units. The prediction is particularly attractive to us
1. Introduction
The construction of molecules that associate in a strong, di-
rectional, and selective mode is a challenging topic in supramo-
lecular chemistry. Because hydrogen bonds are relatively flexible
in geometry as compared to rigid covalent bonds, the principle
of using hydrogen bonds to confer binding strength and se-
lectivity has become an important topic for research.1,2 Many
examples of using hydrogen-bonding interactions to control the
self-assembling of molecules into well-defined aggregates,
ranging from container molecules,3 supramolecular tubes,4 to
* To whom correspondence should be addressed. E-mail: P.-T.C.,
chop@ccms.ntu.edu.tw; M.-k.L., mkleung@ms.cc.ntu.edu.tw; S.-M.P.,
smpeng@mail.ch.ntu.edu.tw.
† National Taiwan University.
‡ Soochow University.
§ Academia Sinica.
| National Taiwan Normal University.
(5) (a) Sakai, N.; Brennan, K. C.; Weiss, L. A.; Matile, S. J. Am. Chem. Soc.
1997, 119, 8726. (b) Sakai, N.; Matile, S. Chem.-Eur. J. 2000, 6, 1731.
(6) MacDonald, J. C.; Whitesides, G. M. Chem. ReV. 1994, 94, 2383.
(7) For recent reviews, see: (a) Piguet, C.; Bernardinelli, G.; Hopfgartner, G.
Chem. ReV. 1997, 97, 2005. (b) Rebek, J., Jr. Chem. Soc. ReV. 1996, 255.
(c) Whitesides, G. M.; Simanek, E. E.; Mathias, J. P.; Seto, C. T.; Chin,
D. N.; Mammen, M.; Gordon, D. M. Acc. Chem. Res. 1995, 28, 37.
(8) For recent examples, see: (a) Corbin, P. S.; Zimmerman, S. C.; Thiessen,
P. A.; Hawryluk, N. A.; Murray, T. J. J. Am. Chem. Soc. 2001, 123,
10475. (b) Archer, E. A.; Sochia, A. E.; Krische, M. J. Chem.-Eur. J.
2001, 7, 2049. (c) So¨ntjens, S. H. M.; Sijbesma, R. P.; van Genderen, M.
H. P.; Meijer, E. W. J. Am. Chem. Soc. 2000, 122, 7487. (d) Archer, E.
A.; Goldberg, N. T.; Lynch, V.; Krische, M. J. J. Am. Chem. Soc. 2000,
122, 5006. (e) Bisson, A. P.; Carver, F. J.; Eggleston, D. S.; Haltiwanger,
R. C.; Hunter, C. A.; Livingstone, D. L.; McCabe, J. F.; Rotger, C.;
Rowan, A. E. J. Am. Chem. Soc. 2000, 122, 8856. (f) Folmer, B. J. B.;
Sijbesma, R. P.; Kooijman, H.; Spek, A. L.; Meijer, E. W. J. Am. Chem.
Soc. 1999, 121, 9001. (g) Sakai, N.; Majumdar, N.; Matile, S. J. Am.
Chem. Soc. 1999, 121, 4294. (h) Sessler, J. L.; Wang, R. Angew. Chem.,
Int. Ed. 1998, 37, 1726. (i) Bisson, A. P.; Carver, F. J.; Hunter, C. A.;
Waltho, J. P. J. Am. Chem. Soc. 1994, 116, 10292. (j) Ghadiri, M. R.;
Granja, J. R.; Milligan, R. A.; McRee, D. E.; Khazanovich, N. Nature 1993,
366, 324.
National Chung Cheng University.
(1) For recent reviews, see: (a) Prins, L. J.; Reinhoudt, D. N.; Timmerman, P.
Angew. Chem., Int. Ed. 2001, 40, 2382. (b) Zimmerman, S. C.; Corbin, P.
S. Struct. Bonding 2000, 96, 63. (c) Niemz, A.; Rotello, V. M. Acc. Chem.
Res. 1999, 32, 44. (d) Ward, M. D. Chem. Soc. ReV. 1997, 26, 365. (e)
Doronina, S. O.; Behr, J. P. Chem. Soc. ReV. 1997, 26, 63. (f) Fyfe, M. C.
T.; Stoddart, J. F. Acc. Chem. Res. 1997, 30, 393.
(2) For examples, see: (a) Corbin, P. S.; Zimmerman, S. C. J. Am. Chem.
Soc. 2000, 122, 3779. (b) Zeng, H.; Miller, R. S.; Flowers, R. A., II; Gong,
B. J. Am. Chem. Soc. 2000, 122, 2635. (c) Appella, D. H.; Barchi, J. J.,
Jr.; Durell, S. R.; Gellman, S. H. J. Am. Chem. Soc. 1999, 121, 2309. (d)
Ranganathan, A.; Pedireddi, V. R.; Rao, C. N. R. J. Am. Chem. Soc. 1999,
121, 1752. (e) Kuduva, S. S.; Craig, D. C.; Nangia, A.; Desiraju, G. R. J.
Am. Chem. Soc. 1999, 121, 1936. (f) Marfurt, J.; Leumann C. Angew.
Chem., Int. Ed. 1998, 37, 175. (g) Coe, S.; Kane, J. J.; Nguyen, T. L.;
Toledo, L. M.; Wininger, E.; Fowler, F. W.; Lauher, J. W. J. Am. Soc.
Chem. 1997, 119, 86.
(3) (a) Castellano, R. K.; Nuckolls, C.; Rebek, J., Jr. J. Am. Chem. Soc. 1999,
121, 11156. (b) Schalley, C. A.; Mart´ın, T.; Obst, U.; Rebek, J., Jr. J. Am.
Chem. Soc. 1999, 121, 2133. (c) Vysotsky, M. O.; Pop, A.; Broda, F.;
Thondorf, I.; Bohmer, V. Chem.-Eur. J. 2001, 7, 4403.
(4) Baumeister, B.; Matile, S. Chem.-Eur. J. 2000, 6, 1739.
9
10.1021/ja011679p CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 4287-4297
4287

A R T I C L E S
Leung et al.
Scheme 1
because a relatively low dissociation energy barrier for the
zipperic dimer would be expected. This implies a fast associa-
tion-dissociation process that may be beneficial for 1 to search
for the optimal matching, resulting in the most stable dimeric
pair.
2. Results and Discussion
2.1. Tactic for the Synthesis of 2-7. Syntheses of 2 and 8
have been reported in the literature.10 Oligomers 3-7, and their
precursors, were prepared on the basis of the Buchwald’s
palladium-catalyzed amination procedures (Scheme 1).11 To
construct the target molecules, terminal building blocks 8 and
the di-tert-butyl substituted 9 were prepared from benzylation
of 2,6-diaminopyridine. Terminal di-tert-butyl substituted 9 was
used as the precursor for 5 and 7 because it could enhance the
solubility12 of 5 and 7. The solubility criterion is particularly
important in the measurements of the dimerization constant.
Compound 10 was obtained from palladium-catalyzed monoam-
To evaluate the dimerization behavior of 1, the dimerization
behavior of its oligomers 2-7 is investigated. Through a
systematic study of the structure of their zipperic hydrogen-
bonded dimers in solid state as well as their thermodynamics
upon hydrogen-bonded dimerization in the solution phase,
valuable information can be provided about the self-dimerization
process. In addition, this study provides a systematic way to
evaluate the Jorgensen’s theory, an important model for the
quantitative prediction of hydrogen-bonding interactions.
(10) (a) Sprinzak, Y. J. Am. Chem. Soc. 1956, 78, 3207. (b) Czuba, W.;
Kowalski, P. Pol. J. Chem. 1980, 54, 853.
(11) (a) Wagaw, S.; Buchwald, S. L. J. Org. Chem. 1996, 61, 7240. (b) Lai,
S.-Y.; Lin, T.-W.; Chen, Y.-H.; Wang, C.-C.; Lee, G.-H.; Yang, M.-h.;
Leung, M.-k; Peng, S.-M. J. Am. Chem. Soc. 1999, 121, 250. (c) Yang,
M.-H.; Lin, T.-W.; Chou, C.-C.; Lee, H.-C.; Chang, H. C.; Lee, G.-H.;
Leung, M-.k.; Peng, S.-M. Chem. Commun. 1997, 2279.
(9) (a) Zimmerman, S. C.; Murray, T. J. Tetrahedron Lett. 1994, 35, 4077;
1995, 36, 7627. (b) Murray, T. J.; Zimmerman, S. C. J. Am. Chem. Soc.
1992, 114, 4010. (c) Pranata, J.; Wierschke, S. G.; Jorgensen, W. L. J.
Am. Chem. Soc. 1991, 113, 2810. (d) Jorgensen, W. L.; Pranata, J. J. Am.
Chem. Soc. 1990, 112, 2008.
(12) Schenk, R.; Gregorius, H.; Meerholz, K.; Heinze, J.; Mu¨llen, K. J. Am.
Chem. Soc. 1991, 113, 2634.
9
4288 J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002

Self-Complementarity of Oligo-2-aminopyridines
A R T I C L E S
Table 1. Crystallographic Data for 2a, 2b, 3, and 4
2a
2b
C19H19N3
3
4
C29H27N7
chemical formula
formula weight
space group
a (Å)
b (Å)
c (Å)
C19H19N3
289.37
P2/c
17.0899(3)
9.0332(1)
15.8559(1)
102.860(1)
2386.38(5)
6
C24H23N5
289.37
P21
381.47
P21/c
473.58
C2/c
12.9834(5)
5.6602(2)
21.9238(7)
100.692(1)
1583.2(1)
4
14.5143(3)
19.6482(4)
14.7183(3)
99.308(2)
4142.1(2)
8
18.7696(4)
12.1599(20
21.8358(5)
98.384(1)
4930.5(2)
8
â (°)
V (Å3)
Z
F(000)
924
616
1616
2000
T (K)
295(2)
0.71073
1.208
0.073
17 892
4217
semiempirical from
equivalents
F 2
295(2)
0.71073
1.214
0.073
6396
295(2)
0.71073
1.223
0.075
25 990
7296
semiempirical from
equivalents
F 2
295(2)
0.71073
1.276
0.079
25 402
4346
semiempirical from
equivalents
F 2
λ (Mo KR) (Å)
D
calc (kg m-3
)
µ (mm-1
)
reflection collected
unique reflections
absorption correction
4529
semiempirical from
equivalents
F 2
refinement on
parameters refined
312
414
548
342
R(Fo)a (I > 2σ(I))
0.0763
0.1785
0.1479
0.2195
1.002
0.0583
0.1067
0.0912
0.1236
1.106
0.0611
0.1188
0.1349
0.1500
1.055
0.0796
0.1410
0.1338
0.1858
1.222
Rw(Fo )b (I > 2σ(I))
2
R(Fo)a (all data)
Rw(Fo ) b(all data)
2
GOF on F 2
2
2
2
2
a R(Fo) ) ∑||Fo| - |Fc||/∑|Fo|. b Rw(Fo ) ) [∑{w(Fo - Fc )2}/∑{w(Fo )2}]1/2
.
Table 2. Geometrical Parameters for N-H‚‚‚N Hydrogen Bonds
Observed in the Crystal Lattices of 2a, 2b, 3, and 4
ination of 2,6-dibromopyridine. The internal building blocks 11
and 12 were obtained from a base-catalyzed coupling reaction
of a 1:1 ratio of 2,6-dibromopyridine and 2,6-diaminopyridine.
Although 12 is expected to be a minor product in our reac-
tion conditions, it could be easily isolated by precipitation in
CH2Cl2 due to the relatively low solubility of 12. The residue
was then concentrated and subjected to liquid chromatography
on silica gel to give 11 as a pure product. Palladium-catalyzed
Buchwald’s coupling of 10 with 11 would afford another
building block 13 in a 38% yield. On the other hand, 12 was
used as an important block for the synthesis of 7. Once the
terminal and the internal building blocks were in hand, oligomers
3-7 were synthesized from the corresponding building blocks
using normal Buchwald’s coupling conditions. However, ad-
dition of 18-crown-6 was required in the synthesis of 6 and 7
to facilitate the coupling reaction.13 Details of syntheses and
spectral characterization are described in the following sections
and in the Experimental Section.
2.2. Single-Crystal X-ray Crystallographic Analyses of
2-4. Single crystals of 2-4 were easily prepared by slow
evaporation of the solvent, and their crystallographic data are
summarized in Table 1. Unfortunately, attempts at preparation
of the single crystals of 5-7 were unsuccessful. Intermolecular
hydrogen-bonding interactions were observed in the crystal
forms of compounds 2-4, of which the hydrogen-bond param-
eters are summarized in Table 2.
crystal
H-bond (Å)
D‚‚‚A (Å)
H‚‚‚A (Å)
D−H‚‚‚A(°)
2a
N(3)-H(3)‚‚‚N(5)
N(4)-H(4)‚‚‚N(2)
mean
3.137(5)
3.008(5)
3.07
2.21(3)
2.20(3)
2.21
160(1)
158(1)
159
2b
3
N(1)-H(1)‚‚‚N(2)
N(4)-H(4)‚‚‚N(5)
mean
N(3)-H(3)‚‚‚N(7)
N(5)-H(5)‚‚‚N(9)
N(6)-H(6)‚‚‚N(2)
N(8)-H(8)‚‚‚N(4)
mean
3.207(5)
3.170(5)
3.19
3.022(5)
3.118(5)
3.071(5)
3.019(5)
3.06
2.54(4)
2.32(4)
2.43
2.08(3)
2.16(4)
2.16(3)
2.09(3)
2.12
146(1)
147(1)
147
178(1)
167(1)
169(1)
177(1)
173
4
N(1)-H(1)‚‚‚N(6)
N(3)-H(3)‚‚‚N(4)
N(5)-H(5)‚‚‚N(2)
mean
3.014(6)
3.025(6)
3.089(6)
3.04
2.15(4)
2.21(4)
2.15(4)
2.17
170(1)
166(1)
166(1)
167
(dashed line). On the other hand, the amino groups of the
central molecule N(4) and N(4A) are bonded to the nitro-
gen atom of pyridine of the other two diaminopyridine mole-
cules N(2) and N(2A) (dashed line). The N‚‚‚N distances are
found to be 3.173(5) and 3.008(5) Å for N(3)‚‚‚N(5) and
N(4)‚‚‚N(2) hydrogen bonds, respectively. The hydrogen-bond
H‚‚‚N distances of 2.21(3) Å for N(3)-H(3)‚‚‚N(5) and
N(3A)-H(3A)‚‚‚N(5) are slightly longer than those of 2.20(3)
Å for N(4)-H(4)‚‚‚N(2) and N(4A)-H(4A)‚‚‚N(2A). Another
3-D picture of 2a depicted in Figure 1b clearly displays the
spatial arrangement among these trimers.
The second form 2b is P21 with Z ) 4 (Figure 2). Two
crystallographic independent, one-dimensional zigzag hy-
drogen-bonding networks are found in this structure. Two chains
are parallel to the b axis with different orientations. In one
chain, one amino group N(1)-H(1) is hydrogen bonded to the
pyridine N(2A) of the next layer. A similar bonding pattern
also occurs in the other chain, in which N(4)-H(4) is hy-
drogen bonded to the pyridine N(5A) of the next layer. The
presence of the hydrogen bonds aligns the benzene ring and
Two crystallization patterns by the molecular aggregation of
2 are observed; one is a trimeric form 2a (Figure 1), and the
other is a zigzag polymeric chain 2b (Figure 2). The 2a form is
P2/c with Z ) 6 in which an interesting trimer sharing four
hydrogen bonds with a 2-fold axis through the N(5), C(22) atoms
is found (Figure 1a). The nitrogen atom N(5) of the central
pyridine is symmetrically hydrogen bonded to two amino
groups N(3) and N(3A) of two other diaminopyridine molecules
(13) Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 1997, 62, 6066.
9
J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002 4289

A R T I C L E S
Leung et al.
bond interactions. In general, a strong hydrogen-bonding
interaction would lead to short NH‚‚‚N and N‚‚‚N distances,
and a large NH‚‚‚N angle.14 The N‚‚‚N distances vary in a
range of 3.019(5)-3.118(5) Å for 3 with an average of 3.06
Å, and 3.014(6)-3.089(6) Å with an average of 3.04 Å for 4.
The H‚‚‚N distances vary in a range of 2.08(3)-2.16(4) Å for
3 with an average of 2.12 Å, and 2.15(4)-2.21(4) Å with an
average of 2.17 Å for 4. These values are slightly larger than
the reported statistical mean distances of 2.939 Å (σ )
0.092) for N‚‚‚N and 2.04 Å (σ ) 0.14) for R2NH‚‚‚N (sp2),
but still fall within the boundary of standard deviations.15 The
average NH‚‚‚N angles of 173° and 167° for 3 and 4,
respectively, in comparison to the statistical mean of 166° (σ
) 8.7) also suggest reasonably strong intermolecular hydrogen-
bond interactions.
The distances between the donor nitrogen atoms and between
the acceptor nitrogen atoms of the proximal hydrogen-bond
donor-acceptor pairs are of particular importance because the
secondary interactions arising from those atoms strongly affect
the stability of the dimers. Data listed in Table 3 show an
average N-N separation of 3.72 Å for the dimer of 3 and 3.69
Å for the dimer of 4. Because these distances are still short,
substantial electrostatic interactions are expected.9c
2.3. Thermodynamics of Dimerization. The possibility of
the formation of hydrogen-bonded dimers in solution was first
evaluated by electrospray ionization mass spectroscopy (ESI-
MS).16 Although mass spectrometry only reflects the properties
of a gas-phase species, it has been known that the correlation
between the gas phase and solution is often good for ESI-MS.
In comparison to 4, the relatively soluble tert-butyl substituted
Figure 1. (a) The molecular structure of 2 in a trimeric form. Ellipsoids
are drawn at the 30% probability level. (b) Perspective view of 2a along
the crystallographic b axis.
pyridine ring of 2 packed in an orderly arrangement (Figure
2b), where the interlayer distance is equal to the length of
the b axis (5.660 Å). The N(1)‚‚‚N(2A) and N(4)‚‚‚N(5A)
distances are found to be 3.207(5) and 3.170(5) Å, respec-
tively. The relatively long hydrogen-bond H‚‚‚N distances
of 2.54(4) and 2.32(4) Å for N(1)-H(1)‚‚‚N(2A) and
N(4)-H(4)‚‚‚N(5A) with small bond angles of 146(1)° and
147(1)° in 2b indicate weaker hydrogen-bonding interactions
between molecules in comparison to those in 2a.14 We
tentatively attributed this observation to the lower hydrogen-
bonding density in 2b, in which only one pair of hydrogen bond
per molecule was formed. However, the existence of two forms
of diaminopyridine in the solid state indicates that the two kinds
of hydrogen-bonding network may only slightly differ in their
free energy.
1
5 is more appropriate for the latter H NMR and UV-vis
studies. Thus, among the list of compounds, we chose 2, 3, 5,
6, and 7 for the ESI-MS analysis. The signals of m/z ) 2M +
H in the ESI-MS analysis (Table 4) support the formation of
dimers of 2, 3, 5, 6, and 7 in CHCl3. However, the signal
intensities in the ESI-MS measurement may be affected by
many experimental factors, and the relative intensities do not
simply reflect the relative amounts of the species in solution.
1
Therefore, further H NMR (in CDCl3) and UV-vis spectro-
scopic experiments (in cyclohexane) were carried out to
substantiate the MS results. These experiments should provide
quantitative information about the hydrogen-bond directed
zipperic self-dimerization.
1
1H NMR Approaches. H and 13C NMR spectra taken in
Compounds 3 and 4 crystallize in a spiral conformation of
the dimeric form possessing four and six pairs of N-H‚‚‚N
hydrogen bonds, respectively, which are shown in Figures 3
and 4. Triaminodipyridine 3 crystallizes in P21/c with Z ) 8.
There is one helical dimer in an asymmetric unit (Figure 3).
Four pairs of hydrogen bonds form between the amino groups
and the nitrogen atoms of the pyridine groups. Similarly,
tetraaminotripyridine 4 crystallizes in C2/c with Z ) 8. The
helical dimer is located at the crystallographic 2-fold axis in
which only one molecule is found in an asymmetric unit (Figure
4). Six pairs of hydrogen bonds form in the dimeric configu-
ration of 4.
CDCl3 are in agreement with the structural assignments for
compounds 2-7. Figure 5 shows the plot of NH’s chemical
shifts versus concentrations for 2, 3, 5, 6, and 7 at 30 °C. Note
that the NH peak was chosen to be well separated from the rest
of the peaks to gain accuracy. Details for the choice of the
specific NH’s chemical shifts are provided in the Supporting
Information. As shown in Figure 5, the concentration-
dependence of the δ values for the amino protons of 2, 3, 5, 6,
and 7 suggests hydrogen-bonded dimer formation.17 The dimer-
ization constants Kd, δi, and δf were obtained from nonlinear
(15) Llamas Saiz, A. L.; Foces-Foces, C. J. Mol. Struct. 1990, 238, 367.
(16) For example, see: Schalley, C. A.; Castellano, R. K.; Brody, M. S.;
Rudkevich, D. M.; Siuzdak, G.; Rebek, J., Jr. J. Am. Chem. Soc. 1999,
121, 4568.
(17) Connors, K. A. Binding Constants, The Measurement of Molecular Complex
Stability; Wiley: New York, 1987; p 202. To simplify the curve fitting
process, δi of 2, 3, and 5, and δf of 7 could be directly evaluated from the
plot in Figure 5.
The distances between NH‚‚‚N and N‚‚‚N, as well as the
NH‚‚‚N angle, are useful parameters for evaluating the hydrogen-
(14) Jeffrey, G. A. An Introduction to Hydrogen Bonding; Oxford: New York,
1997; p 12.
9
4290 J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002

Self-Complementarity of Oligo-2-aminopyridines
A R T I C L E S
Figure 2. (a) The molecular structure of 2 in a polymeric zigzag form. Ellipsoids are drawn at the 30% probability level. (b) Perspective view of 2b along
the crystallographic b axis.
Figure 4. The dimeric structure of 4. Ellipsoids are drawn at the 30%
probability level.
Figure 3. The dimeric structure of 3. Ellipsoids are drawn at the 30%
probability level.
values for the dimerization free energy are listed in Table 5.
The dimerization constant Kd of 2 is too low to be determined
least-squares fitting of the data with eq 1 expressed as17
accurately. The low Kd value may qualitatively rationalize the
1 + 4K [OAP] - 1 + 8K [OAP]
x
d
d
unfavorable dimeric formation in the crystal packing. Instead,
trimer 2a and zigzag polymeric chain 2b are dominant forms
(vide supra). In CDCl3, a fair estimation of the Kd value for 2
can be made. The magnitude of Kd (0.034 M-1) for 2 in our
estimation is reasonably close to the reported value of Kd (0.2
M-1) for 2,6-(C5H11(CO)NH)2C5H3N.18 In addition, the cor-
relation plot (Figure 6a) shows that the dimerization free energy
δ ) δi +
(δf - δi) (1)
4Kd[OAP]
where [OAP] is the concentration of oligo-2-aminopyridines
used in the measurement. δi and δf are the NH chemical shifts
of oligo-2-aminopyridines in the monomeric and dimeric forms,
respectively. The resulting dimerization constants along with
9
J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002 4291

A R T I C L E S
Leung et al.
Table 3. Distances between the Donor Nitrogen Atoms and the
Acceptor Nitrogen Atoms of the Proximal Hydrogen-Bond
Donor-Acceptor Pairs (in Å)
dimer of 3
dimer of 4
N1‚‚‚N6
N2‚‚‚N7
N3‚‚‚N6
N3‚‚‚N8
N4‚‚‚N7
N4‚‚‚N9
N5‚‚‚N8
N5‚‚‚N10
mean
3.798
3.677
3.674
3.747
3.647
3.704
3.669
3.897
3.72
N1‚‚‚N7
N1‚‚‚N5
N2‚‚‚N6
N2‚‚‚N4
N3‚‚‚N5
N3‚‚‚N3
N4‚‚‚N4
3.788
3.628
3.763
3.665
3.628
3.547
3.804
mean
3.69
Table 4. Electrospray Ionization Mass Spectral Data of 2, 3, 5, 6,
and 7
Figure 6. Plot of ∆G0dim (kJ mol-1) versus n, the number of repeating
2-aminopyridine units: (a) in CDCl3 (303 K), (b) in cyclohexane (298 K).
m/z ) M + H
(base peak)
m/z ) 2M + H
(rel intensity %)
compounda
stabilization free energy arising from the hydrogen-bonding
interactions per repeating unit, and P and S are the primary and
secondary interactions defined by Jorgensen, respectively.
However, our correlation plot in Figure 6a shows an empirical
correlation of ∆G0dim ) n(∆∆Gd0im) + ∆G0int which is slightly
different from the predicted equation. Linear least-squares fitting
of the data gives ∆∆Gd0im ) -8.0 kJ/mol (the slope) and ∆G0int
) 17.3 kJ/mol (the intercept). Although as foreseen by Jor-
gensen’s model where the dimerization free energy is linearly
proportional to the number of the repeating units, the presence
of a positive intercept, ∆G0int,20 in the correlation plot should
not be ignored. Incorporating Schneider’s empirical values of
-7.9 and 2.9 kJ mol-1 for the primary attractive and the
secondary repulsive interactions,19b respectively, an increment
of ∆∆G0dim ) -4.2 kJ/mol for each of the additional repeating
units is anticipated from Jorgensen’s equation. In comparison
with this prediction, an increment of ∆∆G0dim ) -8.0 kJ/mol
from our experiments is larger than the predicted one. These
results indicate that the hydrogen-bonding interactions between
our helical dimeric pairs are stronger than Schneider’s statistical
average. These results also suggest that mismatched pairing in
the dimer formation is less favored. Mismatched pairing would
lead to an increase of the ∆G0dim, leading to a less stable
dimeric form. The presence of a positive intrinsic free energy,
∆G0int, indicates the existence of intrinsic factors that prevent
the monomers from dimerization. These factors are independent
of the hydrogen-bonding interactions in dimers, regardless of
the units’ number. Several possible factors including steric
repulsion between terminal substituents, solvation effects, and
entropy loss in dimerization may contribute to the nonnegligible
∆G0int.21
2
3
5
6
7
290.2
382.3
689.6
566.4
882.9
577.3(4)
762.7(1)
1396.2(19)
1130.9(5)
1764.0(3)
a The concentrations of the samples were ranged from 1 × 10-3 to 3 ×
10-3 M.
Figure 5. The plot of NH’s chemical shifts versus concentrations for 2
(9), 3 (2), 5 (1), 6 ([), and 7 (b) at 30 °C. (-) The best fitted curves for
each plot using eq 1 (see text).
Table 5. Dimerization Constants and Thermodynamic Parameters
(kJ mol-1) for 2, 3, 5, 6, and 7 in CDCl3 at 303 K
c
c
0
0
dim
0
compound
n
K [M-1
]
∆G
∆H
T∆S
dim
d
dim
2
3
5
6
7
1
2
3
4
5
3.4 × 10-2 a
5.5 × 10-1 b
6.8b
8.5 ( 2.0
1.5 ( 0.1 -31 ( 3
-4.8 ( 0.2 -47 ( 2
-33 ( 3(32)d
-42 ( 2(41)d
3.4 × 102 b
-14.7 ( 0.6 -75 ( 14 -60 ( 13(59)d
-23.2 ( 0.4
1.0 × 104 b
Temperature variation NMR experiments for 3, 5, and 6 were
carried out to resolve ∆G0dim into ∆H0dim and T∆Sd0im (see Table
5). Linear least-squares fitting of the data gives ∆∆H0dim and
T(∆∆S0dim) to be -22 and -14 kJ/mol, respectively, at 303 K.
Here the negative ∆∆H0dim represents the hydrogen-bonding
dimerization enthalpy per repeating unit, while the negative
T(∆∆S0dim) represents the entropic free-energy loss per repeat-
ing unit due to dimerization. Several different factors such as
a Standard deviation is (2.3 × 10-2 on the basis of nonlinear least-
squares fitting. b Estimated relative error <20%. c The values were obtained
from a linear plot of 1/T versus R ln(Kd), giving the slope equal to ∆H0
dim
and the intercept equal to ∆S0dim d Values of T∆Sd0im at 298 K.
.
∆G0dim is linearly correlated to the number of the repeating
units with r ) 0.99 in CDCl3. On the basis of Jorgensen’s
analysis,9c,d a linear correlation equation of ∆G0dim ) n(2P +
4S) ) n(∆∆G0dim) for our zipperic dimers is expected,19 where
n is the number of repeating units, ∆∆G0dim denotes the
(20) Our definition of the intrinsic free energy ∆G0int in this article is different
from the definition of intrinsic binding energy used by Jencks. For reference,
see: Jencks, W. P. Proc. Natl. Acad. Sci. U.S.A. 1981, 78, 4046.
(21) (a) Williams, D. H.; Searle, M. S.; Mackay, J. P.; Gerhard, U.; Maplestone,
R. A. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 1172. (b) Williams, D. H.;
Searle, M. S.; Groves, P.; Mackay, J. P.; Westwell, M. S.; Beauregard, D.
A.; Cristofaro, M. F. Pure Appl. Chem. 1994, 66, 1975.
(18) Beijer, F. H.; Sijbesma, R. P.; Vekemans, J. A. J. M.; Meijer, E. W.;
Kooijman, H.; Spek. A. L. J. Org. Chem. 1996, 61, 6371.
(19) The equation was expressed in the convention proposed by Schneider. For
references, see: (a) Lu¨ning, U.; Ku¨hl, C. Tetrahedron Lett. 1998, 39, 5735.
(b) Sartorius, J.; Schneider, H.-J. Chem.-Eur. J. 1996, 2, 1446.
9
4292 J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002

Self-Complementarity of Oligo-2-aminopyridines
A R T I C L E S
Table 6. Values of ∆G0dim, ∆Hd0im, and T∆S0dim upon Dimerization
as a Function of the Repeating Unit of 2-Aminopyridine at 298 K
(in kJ mol-1
)
0
dim
0
dim
0
compound
n
∆G
∆H
T∆S
dim
2
3
5
7
1
2
3
5
-(7.2 ( 0.7)
-(15.0 ( 1.5)
-(23.3 ( 2.1)
-(39.6 ( 3.2)
-(52 ( 3)
-(68 ( 5)
-(86 ( 5)
-(124 ( 8)
-(45 ( 3)
-(53 ( 5)
-(63 ( 7)
-(84 ( 10)
prevalently. A detailed derivation of eq 2 is described in the
Appendix. The plot of A0/(A - A0) at, for example, 355 nm as
a function of 1/( A-A ) shown in the insert of Figure 7
x
0
reveals sufficiently linear behavior, supporting the assumption
of dimeric formation in 5. The best linear least-squares fit to
the insert of Figure 7 gives a Kd value of (1.2 ( 0.5) × 104
M-1 (ꢀ3D55 ≈ 8.02 × 104 M-1 cm-1). The hydrogen-bonded
association in 5 can be further supported by the concentration-
dependent fluorescence spectra of which the fluorescence
maximum is gradually red shifted from ∼368 nm (4.2 × 10-6
M) to ∼378 nm (8.4 × 10-5 M, not shown here).23 Dual
fluorescence lifetime of 3.1 and 1.8 ns originating from
monomer and dimer, respectively, was resolved.23 Temperature-
dependent absorption titration spectra were also performed to
resolve ∆Gd0im into ∆Hd0im and T∆Sd0im. The logarithm plot of
dimerization constants as a function of the reciprocal of the
temperature renders linear behavior, and ∆H0dim was deduced
to be ∼-86 kJ/mol for 5. Similar spectroscopic methods were
applied to 2, 3, and 7 in cyclohexane where prevalent hydrogen-
bonded dimerization upon association was also observed.
Accordingly, the best fit of data based on eq 1 gives rise to Kd
values of (1.8 ( 0.5) × 101, (4.2 ( 1.5) × 102, and (8.7 ( 1.5)
× 106 M-1 for 2, 3, and 7,24 respectively. The sparse solubility
of 6 in cyclohexane due to the lack of tert-butyl substituents
makes the extraction of thermodynamic data upon dimerization
infeasible. Figure 6b reveals a plot of the dimerization free
energy ∆G0dim as a function of the number of the repeating
2-aminopyridine units n. Again, as foreseen by the Jorgensen
model, the dimerization free energy is linearly proportional to
the number of the repeating hydrogen-bond units. A linear least-
squares fit of the data based on an empirical correlation of
∆G0dim ) n(∆∆G0dim) + ∆Gi0nt results in ∆∆Gd0im and ∆Gi0nt of
-8.3 and 1.5 kJ/mol, respectively. The fact that the value of
∆G0int in cyclohexane is much smaller than that in CDCl3,
indicating the applicability of the Jorgensen model in a nonpolar
solvent, is worth mentioning here. Table 6 lists the values of
∆G0dim, ∆H0dim, and T∆Sd0im (298 K) upon dimerization as a
function of the repeating unit of 2-aminopyridine in cyclohexane.
The correlation plots of the data obtained from temperature
variation experiments versus the number of the repeating units
result in ∆∆H0dim ) -18 kJ/mol and T(∆∆Sd0im) ) -9.8 kJ/mol
at 298 K (not shown here).
Figure 7. The concentration-dependent absorption spectra of 5 in cyclo-
hexane, in which C0 was prepared to be (a) 4.2 × 10-6, (b) 8.4 × 10-6, (c)
1.68 × 10-5, (d) 3.36 × 10-5, (e) 6.72 × 10-5 M. The spectra are
normalized at 345 nm. Insert: The plot of A0/(A - A0) at, e.g., 355 nm as
a function of 1/(A-A ).
x
0
internal rotational restrictions of monomers due to dimer
formation or desolvation of the solvent molecules may have
contributed to T(∆∆S0dim). Even though in principle the slope
and intercept of the plots to n ) 0 would give ∆H0int and
T∆S0int, respectively, uncertainty due to curve fitting would
limit the reliability of these parameters, leading to a less
conclusive prediction.
UV-Vis Absorption Measurements. It is of key importance
to mimic the intrinsic hydrogen-bonding effect in the nonpolar
solvent that provides an environment free from the solvent
perturbation such as polar(solute)-polar(solvent) and/or solvent
hydrogen-bond interactions. Alternatively, the concentration-
dependent absorption spectroscopy offers a reliable method
because the dual hydrogen-bond formation of the 2-aminopy-
ridine unit might alter the S0 f S1 (π, π*) chromophore due to
the associated charge-transfer effect.22 Here, compound 5 was
used as a prototype to demonstrate such feasibility. When the
concentration was prepared to be as low as 1.2 × 10-6 M, 5
exhibits the lowest singlet π f π* absorption band maximum
at 345 nm (ꢀ345 ≈ 4.3 × 104 M-1 cm-1). Upon increasing the
concentration, changes in the spectral features were observed
in which the spectra after normalization reveal significant red
shift (see Figure 7a-e). The results unambiguously conclude
the formation of dimer and/or higher-order aggregates through
the hydrogen-bonding effect. On the basis of the dimeric
structure resolved from X-ray, we first assume a dominant
dimeric formation, which possesses a spiral, cyclic type of six
hydrogen-bond configurations. The dimerization constant Kd can
thus be extracted from eq 2
A0
lꢀ2M
2ꢀM
1
1
)
‚
‚
+
(2)
A - A0
ꢀ - 2ꢀ
K
A - A
ꢀD - 2ꢀM
x
x x
D
M
d
0
The enthalpy/entropy compensation effect21,25 is observed in
our system. Figure 8 shows a plot of ∆H0dim versus T∆Sd0im at T
) 298 K that integrates the data for 2, 3, 5, 6, and 7 in both
solvents. The plot is a sufficiently straight line with a slope of
where A and l denote the initially prepared absorbance at a
selected wavelength and cell path length, respectively. ꢀM and
ꢀD are the molar extinction coefficients of monomer and dimer
at the selected wavelength. A0 is the absorbance of the monomer
assuming that no dimer is formed. Experimentally, ꢀM can be
resolved by performing the absorption titration study at a
sufficiently low concentration where the monomer exists
(23) To avoid the inner filter effect, a configuration of front-face excitation was
applied throughout the fluorescence study.
(24) Because of the large Kd value, a very low concentration of 7 is required to
perform the concentration-dependent study. Accordingly, 4 cm and 10 cm
path length cells were incorporated in the absorption titration study.
(25) Williams, D. H.; O’Brien, D. P.; Bardsley, B. J. Am. Chem. Soc. 2001,
123, 737.
(22) Chou, P. T.; Wu, G. R.; Wei, C. Y.; Cheng, C. C.; Chang, C. P.; Hung, F.
T. J. Phys. Chem. B 2000, 104, 7818.
9
J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002 4293

A R T I C L E S
Leung et al.
preliminary approach should provide certain clues about the
relative importance of solute-solute and solute-solvent interac-
tions.
Several interesting observations were made from the calcula-
tion results. First, all calculated (C-)H‚‚‚N distances are within
the sum of van der Waals radii of H and N (2.6 Å), while all
(N-)H‚‚‚Cl and (C-)H‚‚‚Cl distances are larger than the sum
of van der Waals radii of H and Cl (2.9 Å). In almost all cases,
the C-H‚‚‚N angle is more linear than the N-H‚‚‚Cl and
C-H‚‚‚Cl angles. The results implied that C-H‚‚‚N interaction
is of primary importance, while the N-H‚‚‚Cl and C-H‚‚‚Cl
interactions are minor parts. Second, at the B3LYP/6-31+G**//
HF/6-31G* level, the gas-phase complexation energies of
2-Me‚CHCl3 (-17.45 kJ/mol) and 3-Me‚2CHCl3 (-23.47 kJ/
mol) are a few kJ/mol smaller than those of (2-Me)2 (-28.87
kJ/mol) and (3-Me)2 (-42.63 kJ/mol). Thus, the magnitude of
the solute-solvent interaction in chloroform is far from
negligible. In comparison, much less solvation outcomes can
be expected with less polar or acidic solvents such as cyclo-
hexane, consistent with the experimental results. Finally, the
dimerization energies of 2-Me, 3-Me, and 4-Me were calculated
to be -28.87, -42.63, and -61.63 kJ/mol, respectively, at the
B3LYP/6-31+G**//HF/6-31G* level. The plot of ∆Edimer (see
Experimental Section) versus the number of the repeating
2-aminopyridine groups shows linear behavior (R2 ≈ 0.991, not
shown here). However, the increase of complexation energy is
not a simple multiple of the number of hydrogen bonds. For
example, the complexation energy of (3-Me)2 is less than
twice that of (2-Me)2, possibly due to secondary interactions.
In (2-Me)2, the unbound N-H group twists out of the molecular
plane to avoid secondary interactions. However, for the forma-
tion of two additional hydrogen bonds in (3-Me)2, the middle
N-H group, which corresponds to the unbound N-H bond in
(2-Me)2, has to be brought back to the vicinity of a neighboring
N-H group of the binding partner (see Figures 2 and 3 for
reference). For 3-Me‚2CHCl3, the unfavorable dipole-dipole
interaction between the two chloroform molecules cancels out
some of the favorable solute-solvent interactions. Therefore,
decomplexation of the second chloroform molecule becomes
relatively easy. A similar trend was applied to oligo-2-
aminopyridines with higher repeating n (>3) units. The results
qualitatively rationalize the difference in ∆Gint, while ∆∆Gdim
is similar between CDCl3 and cyclohexane. Further detailed
studies focusing on the molecular dynamics simulation are in
progress.
Figure 8. The correlation plot of ∆H0dim (kJ mol-1) versus T∆S0dim (kJ
mol-1) at 298 K: (2) data in CDCl3; (9) data in cyclohexane.
0.54, suggesting that around 54% of the enthalpy gain from
dimerization interactions would be sacrificed to compensate for
the entropic free-energy loss. On the basis of the plot, one could
perceive that large decreases in enthalpies are linearly correlated
to the decreases in entropies upon dimerization. Data obtained
either in cyclohexane or in CDCl3 follow the same correlation.
These results could be rationalized by the hydrogen-bond effect.
Although the hydrogen-bonding formation between the mono-
mers would gain significant amounts of the enthalpic stabiliza-
tion energy, the increased hydrogen-bond interactions would
reduce internal motions of the monomers, resulting in a loss of
entropy. These results also imply the formation of a relatively
tighter dimeric pair in cyclohexane than that in CDCl3.
The intrinsic CDCl3-solute interaction may be qualitatively
rationalized through an ab initio approach. To simplify the
calculations, the terminal N-benzylic groups were replaced with
N-methyl groups for 2-4, forming 2-Me, 3-Me, and 4-Me. The
initial geometry of monomers and dimers was adopted from
X-ray structures for 3 and 4. As C-H‚‚‚N interactions have
been documented in the literature,26-28 we first attempted to
explore whether chloroform desolvation plays a role in the
process of dimerization. Although the C-H bond of chloroform
has been shown to be a potential proton donor,29-32 the proton
acceptor ability of a chlorine atom is less obvious. Meanwhile,
the interactions between chloroform molecules are quite weak.
The complexation energy of the best geometrically optimized
chloroform dimer that was located is -1.55 kJ/mol and can
thus be neglected. Accordingly, as a crude approximation, it
was assumed that prior to the dimerization formation at least
one C-H‚‚‚N interaction has to be broken for each pyridine
ring. In other words, one and two chloroform molecules have
to be decomplexed from monomeric 2-Me and 3-Me, respec-
tively, in advance of dimerization. Note that more solvent
molecules in the solvation shell have to be relocated in real
solution upon proceeding to dimerization. Nevertheless, our
3. Conclusion
In summary, we have demonstrated a systematic approach
to evaluate the strength of hydrogen-bonding interactions.
Jorgensen’s model can be successfully applied to our systems
in cyclohexane. However, the intrinsic free energy ∆G0int has to
be considered if the dimerization studies are carried out in CDCl3
in which ∆G0int is an important parameter that works against
the monomers from dimerization. Our results show the enthalpy/
entropy compensation relationship as a competition between
bonding and dynamics. Although increasing the number of
hydrogen-bonding interactions would definitely enhance the
hydrogen-bonding stabilization enthalpy, this stabilization en-
thalpy per unit would be partially sacrificed to compensate for
the entropic free-energy loss due to dimerization in our cases.
(26) (a) Desiraju, G. R. Angew. Chem., Int. Ed. Engl. 1995, 34, 2311-2327
and references therein. (b) Marjo, C. E.; Bishop, R.; Craig, D. C.; Scudder,
M. L. Eur. J. Org. Chem. 2001, 863.
(27) Hilfiker, M. A.; Mysak, E. R.; Samet, C.; Maynard, A. J. Phys. Chem. A
2001, 105, 3087.
(28) Gu, Y.; Kar, T.; Scheiner, S. J. Mol. Struct. 2000, 552, 17-31.
(29) (a) Desiraju, G. R. J. Chem. Soc., Chem. Commun. 1989, 179. (b) Desiraju,
G. R. Acc. Chem. Res. 1991, 24, 290-296.
(30) Jemmis, E. D.; Giju, K. T.; Sundararajan, K.; Sankaran, K.; Vidya, V.;
Viswanathan, K. S.; Leszczynski, J. J. Mol. Struct. 1999, 510, 59.
(31) Our calculation at the MP2/6-31G** level estimated the BSSE corrected
binding energy of CHCl3‚‚‚NH3 complex to be -23.26 kJ/mol.
(32) Pawelka, Z.; Koll, A.; Zeegers-Huyskens, Th. J. Mol. Struct. 2001, 597,
57.
9
4294 J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002

Self-Complementarity of Oligo-2-aminopyridines
A R T I C L E S
2,6-amine (9) (0.622 g, 2 mmol) and 2,6-dibromopyridine (0.237 g, 1
mmol) in the presence of Pd2(dba)3 (0.021 g, 2 mol %), BINAP (0.025
g, 4 mol %), and KO-tBu (0.448 g, 4 mmol) in benzene (20 mL) for
8 h gave a crude 5. Purification by column chromatography over silica
gel, using hexane/EtOAc (7:3) as eluent, gave pure 5 (0.468 g, 67%
yield) as yellow solid. Rf 0.50 (hexane/EtOAc, 4:1). mp 120 °C. IR
(KBr): 3409, 3229, 1614 cm-1. 1H NMR (400 MHz, CDCl3): δ 8.75
(s, 2H, NH), 7.44-7.21 (m, 9H), 7.11 (d, J ) 7.6 Hz, 2H), 6.59 (d, J
) 8.0 Hz, 2H), 5.96 (d, J ) 8.0 Hz, 2H), 5.25 (bs, 2H, NH), 4.40 (d,
J ) 5.6 Hz, 4H), 1.31 (s, 36H). 13C NMR (100 MHz, CDCl3): δ 158.53,
153.67, 153.52, 150.95, 139.16, 139.05, 138.18, 121.54, 121.08, 103.24,
99.73, 97.72, 47.27, 34.77, 31.42. FAB-MS: 697 [M+]. Anal. Calcd
for C45H59N7: C, 77.42; H, 8.53; N, 14.05. Found: C, 77.46; H, 8.60;
N, 13.91.
These results clearly support the importance of “the concept of
preorganization”33-35 in the design of hydrogen-bonding guest-
host molecules.
4. Experimental Section
4.1. Materials. Tris(dibenzylideneacetone)dipalladium(0) (Pd2(dba)3,
2,2′-bis(diphenylphosphino)-1,1′-binaphthyl (BINAP), potassium-tert-
butoxide (KO-tBu), and 18-crown-6 were obtained commercially
(Aldrich, Janssan, Tokyo Kasei) and used as received. Benzene was
refluxed over calcium hydride for 8 h before use.
4.2. General Procedures for Synthesizing 2-7. A mixture of the
corresponding bromide and amine, Pd2(dba)3, BINAP, KO-tBu, and
18-crown-6 in dry benzene (10 mL) was refluxed at 80 °C under argon
with constant stirring. After reacting for a specified time period, the
reaction mixture was cooled and quenched with NH4Cl solution (2 mL).
The product was extracted twice with CH2Cl2. The extracts were
combined and washed with H2O (2 mL), dried over anhydrous MgSO4,
and concentrated under reduced pressure to provide a crude oil, which
was further purified by column chromatography over silica gel. In most
cases, a mixture of ethyl acetate and hexane is a suitable eluent for
column chromatography. Although the products isolated by this
approach are reasonably good in their NMR spectra and elemental
analysis, the products usually contained trace amounts of brown
impurity that would significantly affect the experiments in the UV-
vis analysis. Therefore, compounds 2, 3, 5, and 7 were further purified
through column chromatography (diethyl ether) to remove the brown
impurity before analysis.
Bis{6-[6′-(benzylamino)pyrid-2′-yl)amino]pyrid-2-yl}amine (6).
The reaction of N 6-(6′-bromopyrid-2-yl)-N 2-(6′′-(benzylamino)pyrid-
2-yl)pyridine-2,6-diamine (13) (0.446 g, 1 mmol) and (N-benzyl)-
pyridine-2,6-diamine (8) (0.199 g, 1 mmol) in the presence of Pd2(dba)3
(0.021 g, 2 mol %), BINAP (0.025 g, 4 mol %), KO-tBu (0.448 g, 4
mmol), and 18-crown-6 (0.65 g, 2.5 mmol) in benzene (10 mL) for 8
h gave a crude 6. Purification by column chromatography over silica
gel (CH2Cl2/acetone, 24:1) gave pure 6 (0.362 g, 64% yield) as a brown
solid. Rf 0.38 (hexane:EtOAc, 3:2). mp 181 °C. IR (KBr): 3421, 3195,
1602, 1580 cm-1. 1H NMR (400 MHz, CDCl3): δ 10.17 (s, 1H, NH),
9.73 (s, 2H, NH), 7.41 (t, J ) 8 Hz, 2H), 7.29-7.18 (m, 12H), 6.99
(d, J ) 8 Hz, 2H), 6.82 (d, J ) 8 Hz, 2H), 6.53 (d, J ) 8 Hz, 2H),
5.86 (d, J ) 8 Hz, 2H), 5.62 (bs, 2H, NH), 4.35 (d, J ) 5.6 Hz, 4H).
13C NMR (100 MHz, DMSO-d6): δ 157.45, 153.12, 152.68, 152.58,
140.93, 138.37, 138.11, 128.14, 127.06, 126.40, 102.92, 102.86, 98.92,
98.72, 44.35. FAB-MS: 565 [M+]. HRMS calcd for C34H31N9,
565.2702; found, 565.2698.
N 2,N6-{6′-[6′′-(3′′′,5′′′-Di-tert-benzylamino)pyrid-2′′-yl)amino]py-
rid-2′-yl}pyridine-2,6-diamine (7). The reaction of N-(3′,5′-di-tert-
butylbenzyl)pyridine-2,6-amine (9) (0.622 g, 2 mmol) and N 2,N6-bis(6′-
bromopyrid-2′-yl)pyridine-2,6-diamine (12) (0.419 g, 1 mmol) in the
presence of Pd2(dba)3 (0.032 g, 3 mol %), BINAP (0.038 g, 6 mol %),
KO-tBu (0.56 g, 5 mmol), and 18-crown-6 (0.528 g, 2 mmol) in benzene
(10 mL) for 8 h gave a crude 7. Purification by column chromatography
over silica gel (hexane/EtOAc, 3:2) gave pure 7 (581 mg, 66%) as a
yellow solid. Rf 0.58 (hexane:EtOAc, 7:3). mp 139 °C. IR (KBr): 3418,
3228, 3201, 1613 cm-1. 1H NMR (400 MHz, CDCl3): δ 11.39 (s, 2H,
NH), 10.70 (bs, 2H, NH), 7.44-7.40 (m, 3H), 7.20-7.36 (m, 6H),
7.19 (s, 4H), 7.07 (bs, 2H), 6.76 (m, 4H), 6.46 (d, J ) 8.0 Hz, 2H),
5.90 (d, J ) 8 Hz, 2H), 4.38 (bs, 4H), 1.23 (s, 36H). 13C NMR (100
MHz, CDCl3): δ 154.65, 154.61, 154.39, 153.92, 151.01, 150.98,
139.33, 139.25, 139.16, 138.16, 121.24, 121.01, 103.68, 103.14, 102.86,
99.75, 97.56, 47.33, 34.78, 31.43. FAB-MS: 881 [M+]. Anal. Calcd
for C55H67N11: C, 74.87; H, 7.66; N, 17.47. Found: C, 74.06; H, 7.62;
N, 16.96.
Bis[6-(benzylamino)pyrid-2-yl]amine (3). The reaction of (N-
benzyl)pyridine-2,6-diamine (8) (0.199 g, 1 mmol) and 2-(N-benzyl-
amino)-6-bromopyridine (10) (0.262 g, 1 mmol) in the presence of Pd2-
(dba)3 (0.010 g, 1 mol %), BINAP (0.012 g, 2 mol %), and KO-tBu
(0.448 g, 4 mmol) in benzene (10 mL) for 8 h gave a crude product 3.
Purification by column chromatography over silica gel, using hexane/
EtOAc (4:1) as eluent, gave the pure bis[6-(benzylamino)pyrid-2-yl]-
amine (3) (0.304 g, 79% yield) as colorless crystals. Rf 0.50 (hexane/
EtOAc, 7:3). mp 131 °C. IR (KBr): 3421, 3265, 1610 cm-1. 1H NMR
(200 MHz, DMSO-d6): δ 8.47 (s, 1H, NH), 7.36-7.09 (m, 12H), 6.77
(d, J ) 8.0 Hz, 2H), 6.73 (t, J ) 6 Hz, 2H, NH), 5.93 (d, J ) 8.0 Hz,
2H), 4.46 (d, J ) 6 Hz, 4H). 13C NMR (100 MHz, DMSO-d6): δ
158.89, 139.63, 139.43, 139.37, 128.19, 127.27, 126.65, 114.05, 106.53,
44.18. EI-MS: 381 [M+]. Anal. Calcd for C24H23N5: C, 75.55; H, 6.08;
N, 18.37. Found: C, 75.19; H, 6.15; N, 18.86.
N 2,N 6-Bis[6′-(benzylamino)pyrid-2′-yl]pyridine-2,6-diamine (4).
The reaction of (N-benzyl)pyridine-2,6-diamine (8) (0.398 g, 2 mmol)
and 2,6-dibromopyridine (0.237 g, 1 mmol) in the presence of Pd2-
(dba)3 (0.021 g, 2 mol %), BINAP (0.025 g, 4 mol %), KO-tBu (0.56
g, 5 mmol), and 18-crown-6 (1.3 gm, 5 mmol) in benzene (20 mL) for
16 h gave a crude 4. Purification by column chromatography over silica
gel, using hexane/EtOAc (7:3) as eluent, gave 4 (0.34 g, 72% yield) as
yellow crystals. Rf 0.45 (hexane/EtOAc, 7:3). mp 188 °C. IR (KBr):
N-(3′,5′-Di-tert-butylbenzyl)pyridine-2,6-diamine (9). To a mixture
of 2,6-diaminopyridine (1.09 g, 0.01 mol) and KO-tBu (2.24 g, 0.02
mol) in dry benzene (30 mL) was added a solution 3,5-di-tert-
butylbenzylbromide (2.83 g, 0.01 mol) in benzene (5 mL) at room
temperature with constant stirring. The mixture was refluxed at 80 °C
for 6 h. After cooling, the crude mixture was quenched with saturated
solution of NH4Cl (4 mL), washed with water (4 mL), and dried over
anhydrous MgSO4. The solvent was removed, and the crude product
was chromatographed over silica gel (hexane/EtOAc, 3:1) to give 9
(2.36 g, 76%) as a light brown thick liquid. Rf 0.32 (hexane/EtOAc,
1
3437, 3232, 1604 cm-1. H NMR (300 MHz, DMSO-d6): δ 8.67 (s,
2H, NH), 7.37-7.20 (m, 13H), 7.12 (d, J ) 7.8 Hz, 2H), 6.85-6.81
(m, 4H), 6.00 (d, J ) 7.8 Hz, 2H), 4.49 (d, J ) 6 Hz, 4H). 13C NMR
(100 MHz, DMSO-d6): δ 157.45, 153.14, 152.68, 140.97, 138.14,
138.09, 128.14, 127.07, 126.40, 102.64, 98.90, 98.72, 44.32. FAB-
MS: m/z 473 [M+]. Anal. Calcd for C29H27N7: C, 73.53; H, 5.74; N,
20.71. Found: C, 73.20; H, 5.73; N, 20.73.
N2,N6-Bis[6′-(3′′,5′′-di-tert-butylbenzylamino)pyrid-2′-yl]pyridine-
2,6-diamine (5). The reaction of N-(3′,5′-di-tert-butylbenzyl)pyridine-
1
4:1). IR (KBr): 3476, 3380, 3302, 1609 cm-1. H NMR (400 MHz,
CDCl3): 7.44 (t, J ) 1.6 Hz, 1H), 7.23-7.30 (m, 3H), 5.85-5.88 (m,
2H), 4.96 (t, J ) 5.6 Hz, 1H), 4.18-4.20 (m, 4H), 1.41 (2, 18H). 13C
NMR (100 MHz, CDCl3): 31.29, 34.59, 47.03, 95.16, 96.79, 120.88,
121.65, 138.15, 139.12, 150.71, 157.57, 158.12. MS: m/z (EI 70 eV)
311 (M+, 100). Anal. Calcd for C20H29N3: C, 77.11; H, 9.39; N, 13.50.
Found: C, 76.83; H, 9.29; N, 13.34.
(33) Cram, D. J. Angew. Chem., Int. Ed. Engl. 1986, 25, 1039.
(34) Martell, A. E.; Hancock, R. D.; Motekaitis, R. J. Coord. Chem. ReV. 1994,
133, 39.
(35) For recent examples based on the concept of preorganization, see: (a) Bell,
T. W.; Khasanov, A. B.; Drew, M. G. B.; Filikov, A.; James, T. L. Angew.
Chem., Int. Ed. 1999, 38, 2543. (b) Bell, T. W.; Hou, Z.; Zimmerman, S.
C.; Thiessen, P. A. Angew. Chem., Int. Ed. Engl. 1995, 34, 2163.
9
J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002 4295

A R T I C L E S
Leung et al.
Preparation of N2,N6-Dibenzylpyridine-2,6-diamine (2) and 2-N-
Benzylamino-6-bromopyridine (10). To a magnetically stirred solution
of 2,6-dibromopyridine (2.37 g, 0.01 mol), KO-tBu (1.40 g, 0.0125
mol), Pd2(dba)3 (0.025 mg, 0.25 mol %), and BINAP (30 mg, 0.5 mol
%) in dry benzene (40 mL) was added benzylamine (1.09 mL, 0.01
mol) dropwise at room temperature under argon atmosphere. The
reaction mixture was refluxed at 80 °C for 4 h. After cooling, the
reaction mixture was quenched with saturated NH4Cl solution
(4 mL), washed with H2O (4 mL), and dried over anhydrous
MgSO4. The solvent was removed under vacuum, and purification by
column chromatography over silica gel (hexane/EtOAc, 19:1) gave
2-benzylamino-6-bromopyridine (10) (575 mg, 22%) as colorless
crystals. Rf 0.47 (hexane/EtOAc, 9:1). mp 88 °C. IR (KBr): 3288,
8.90 (s, 1H, NH), 8.17 (d, J ) 8.1 Hz, 1H), 7.54 (t, J ) 7.8 Hz, 1H),
7.17-7.41 (m, 8H), 7.01 (d, J ) 7.5 Hz, 1H), 6.90 (t, J ) 6.0 Hz, 1H,
NH), 6.76 (dd, J ) 6.1 and 2.2 Hz, 1H), 6.58 (d, J ) 7.8 Hz, 1H),
6.00 (d, J ) 8.1 Hz, 1H), 4.48 (d, J ) 6.0 Hz, 2H). 13C NMR (100
MHz, DMSO-d6): 44.40, 98.65, 98.96, 102.89, 103.55, 110.45, 118.46,
126.42, 127.03, 128.16, 138.11, 138.45, 138.68, 140.38, 140.91, 151.97,
152.60, 153.12, 154.49, 157.52. FAB-MS: m/z 448 (M+ + 2), 446
(M+). Anal. Calcd for C22H19BrN6: C, 59.07; H, 4.28; N, 18.79.
Found: C, 58.54; H, 4.24; N, 18.52.
4.3. Methods. 1H and 13C NMR were recorded in CDCl3 or DMSO-
d6, and chemical shifts were reported in ppm relative to CHCl3 (δ 7.24
1
1
ppm for H and 77.0 ppm for 13C) or DMSO-d6 (δ 2.49 ppm for H
and 39.5 ppm for 13C). Melting points were uncorrected. The variable-
concentration 1H NMR measurements (400 MHz) for 2, 3, 5, 6, and 7
were run in CDCl3, using the signal of CHCl3 (δ ) 7.25) as the internal
standard. Before use, CDCl3 was treated with K2CO3 to remove any
acidic impurity, dried over molecular sieves 4 A, and distilled.
Experimental details for the concentration-dependent chemical shifts
are provided in the Supporting Information.
3084, 1606, 1566 cm-1 1H NMR (200 MHz, DMSO-d6): 7.45 (t,
.
J ) 6 Hz, 1H, NH), 7.19-7.32 (m, 6H), 6.63 (d, J ) 8 Hz, 1H),
6.46 (d, J ) 8 Hz, 1H), 4.40 (d, J ) 6.0 Hz, 2H). 13C NMR (100
MHz, DMSO-d6): 44.24, 98.61, 126.32, 127.03, 128.06, 137.86,
140.96, 153.16, 157.38. MS: m/z (EI, 70 eV) 264 (M+ + 2), 262 (M+).
HRMS (M+) calcd for C12H1179BrN2, 262.0105; found, 262.0099. Also
gave N 2,N 6-dibenzylpyridine-2,6-diamine (2) (490 mg, 17%) as color-
less crystals. Rf 0.7 (hexane/EtOAc, 4:1). 1H NMR (400 MHz,
CDCl3): δ 7.41-7.23 (m, 11H), 5.78 (d, J ) 8 Hz, 2H), 4.84 (t, J )
5.6 Hz, 2H, NH), 4.48 (d, J ) 5.6 Hz, 4H). 13C NMR (100 MHz,
CDCl3): δ 157.91, 139.69, 138.93, 128.34, 127.25, 126.82, 95.00,
46.11.
For the electrospray ionization mass spectroscopy (ESI-MS), a
Finnigan LCQ quadruple ion trap mass spectrometer equipped with a
pneumatically assisted electrospray ionization source was used. The
mass spectrometer was operated in the positive ion mode by applying
a voltage of 4.5 kV to the ESI needle. The temperature of the heated
capillary in the ESI source was set at 200 °C. To avoid a space charge
effect, the number of ions present in the trap at one time was regulated
by automatic gain control, which was set at 5 × 107 ions for the full-
mass target. The flow rate of the sheath gas of nitrogen was set at 30
units. Helium was used as the damping gas at a pressure of 10-3 Torr.
Voltage across the capillary and the octapole lenses was tuned by an
automated procedure to maximize signals from the ion of interest in
the mass spectrum.
Preparation of N-(6′-Bromopyrid-2′-yl)pyridine-2,6-diamine (11)
and N 2,N 6-Bis(6′-bromopyrid-2′-yl)pyridine-2,6-diamine (12). To
a mixture of 2,6-diaminopyridine (1 g, 0.009 mol) and 2,6-dibromopy-
ridine (2.17 g, 0.009 mol) in dry benzene (10 mL) was added KO-tBu
(2.06 g, 0.018 mol) at room temperature under N2 atmosphere with
constant stirring. The reaction mixture was refluxed at 80 °C for 16 h.
After cooling, the crude mixture was quenched with saturated NH4Cl
solution (2 mL), washed with H2O (2 mL), and dried over anhydrous
MgSO4. The solvent was removed under vacuum, and the product was
recrystallized from CH2Cl2 to give N 2,N 6-bis(6′-bromopyrid-2′-yl)-
pyridine-2,6-diamine (12) (310 mg, 8%) as a brown solid. Rf 0.68
Steady-state absorption and emission spectra were recorded by a
U-3310 (Hitachi) spectrophotometer and a F4500 (Hitachi) fluorimeter,
respectively. The sample compartment of U-3310A was attached by a
temperature-controlled unit so that temperature variation studies can
be performed within -10 to 80 °C. Because the titration experiment is
critical for interpretation of the hydrogen-bonding association, we have
carefully performed our absorption and fluorescence titration studies
where each piece of data was acquired from an average of three to
five measurements. The sensitivity for the absorption measurement is
approximately 1.5 × 10-4 in absorbance under constant temperature
((0.1 °C) throughout the measurement.
Ab initio calculation was carried out using Gaussian 98.36 Geometry
optimization and frequency analyses were carried out at the HF/6-31G*
level. All stationary points were verified by performing vibrational
analyses. Dimerization energies reported were results of single point
calculations of ∆Edimer ) Edimer - 2Emonomer, with counterpoise correc-
tions for basis set superposition errors (BSSE), at the B3LYP37,38/6-
31+G** level.
1
(CH2Cl2). IR (KBr): 3262, 3256, 3089, 3067, 3031, 1556 cm-1. H
NMR (200 MHz, DMSO-d6): 7.03 (d, J ) 8.0 Hz, 2H), 7.04 (d, J )
8.0 Hz, 2H), 7.57 (t, J ) 7.8 Hz, 2H), 7.58 (t, J ) 7.9 Hz, 2H), 7.84
(d, J ) 8.2 Hz, 1H), 9.84 (s, 2H). 13C NMR (100 MHz, DMSO-d6):
103.9, 110.4, 118.6, 138.4, 139.2, 140.4, 151.8, 154.2. FAB-MS: m/z
419 (M+). Anal. Calcd for C15H11N5Br2: C, 42.79; H, 2.63; N, 16.63.
Found: C, 42.98; H, 2.67; N, 16.43. The remaining crude mixture was
purified by column chromatography over silica gel (CH2Cl2) to give
N-(6′-bromopyrid-2′-yl)pyridine-2,6-diamine (11) (1.36 g, 56%) as a
light brown solid. Rf 0.24 (CH2Cl2). IR (KBr): 3443, 3305, 3198, 1633,
1570 cm-1. 1H NMR (200 MHz, DMSO-d6): 5.76 (bs, 2H), 5.98 (d, J
) 7.0 Hz, 1H), 6.55 (d, J ) 7.8 Hz, 1H), 6.96 (d, J ) 7.6 Hz, 1H),
7.25 (t, J ) 7.8 Hz, 1H), 7.49 (t, J ) 8.0 Hz, 2H), 8.03 (d, J ) 8.3 Hz,
1H), 9.53 (s, 1H). 13C NMR (100 MHz, DMSO-d6): 98.9, 99.8, 110.2,
117.9, 138.3, 138.5, 140.1, 152.4, 154.7, 158.2. FAB-MS: m/z 264
(M+). Anal. Calcd for C10H9N4Br: C, 45.45; H, 3.44; N, 21.22.
Found: C, 45.55; H, 3.47; N, 20.72.
For the crystal structure determinations, data collection was carried
out on a Siemens SMART diffractometer with a CCD detector (Mo
KR radiation, λ ) 0.71073 Å) at room temperature. A preliminary
N-(6′-Bromo-2′-pyridyl)-N′-(6′′-N-benzylamino-2′′-pyridyl)-2,6-
diaminopyridine (13). To a magnetically stirred solution of 2-N-
benzylamino-6-bromopyridine (10) (524 mg, 2 mmol) and N-(6′-
bromopyrid-2′-yl)pyridine-2,6-diamine (11) (264 mg, 1 mmol) in dry
benzene (20 mL) were added KO-tBu (280 mg, 2.5 mmol), Pd2(dba)3
(21 mg, 2 mol %), BINAP (25 mg, 4 mol %), and 18-crown-6 (528
mg, 2 mmol) at room temperature under argon atmosphere. The stirring
was continued at 80 °C for 4 h. After cooling, the reaction mixture
was quenched with saturated NH4Cl solution (4 mL), washed with H2O
(4 mL), and dried over anhydrous MgSO4. The crude mixture was
purified over silica gel to yield 13 (170 mg, 38%) as colorless crystals.
Rf 0.62 (hexane:EtOAc, 7:3). mp 168 °C. IR (KBr): 3288, 3192, 3020,
1603, 1582 cm-1. 1H NMR (300 MHz, DMSO-d6): 9.72 (s, 1H, NH),
(36) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi,
R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,
J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian,
Inc.: Pittsburgh, PA, 1998.
(37) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(38) (a) Lee, C., Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (b) Miehlich,
B.; Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989, 157, 200.
9
4296 J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002

Self-Complementarity of Oligo-2-aminopyridines
A R T I C L E S
orientation matrix and unit cell parameters were determined from three
runs of 15 frames each, each frame corresponding to 0.3° scan in 20 s,
followed by spot integration and least-squares refinement. For each
structure, data were measured using ω scans of 0.3° per frame for 20
s until a complete hemisphere had been collected. Cell parameters were
retrieved using SMART39 software and refined with SAINT on all
observed reflections. Data reduction was performed with the SAINT40
software and corrected for Lorentz and polarization effects. Absorption
corrections were applied with the program SADABS.40 The structures
of 2a, 2b, 3, and 4 are solved by direct methods with the SHELX-9341
program and refined by full-matrix least-squares methods on F 2 with
SHEXLTL-PC V 5.03.42 All non-hydrogen atomic positions were
located in difference Fourier maps and refined anisotropically. Hydrogen
atoms attached to nitrogen atoms were located and refined isotropically.
The rest of the hydrogen atoms attached to carbon atoms were fixed at
calculated positions and refined using a riding mode. Detailed crystal
data of 2a, 2b, 3, and 4 are listed in Table 1, and the bond lengths and
angles are deposited in the Supporting Information (CIF files).
Crystal Data of 2a. Single crystals of 2 were obtained by
recrystallization from hexanes/ethyl acetate. A light-yellow crystal of
approximate dimensions 0.4 × 0.4 × 0.25 mm3 was mounted on a
glass capillary. A total of 17 886 reflections was collected with a final
resolution of 0.75 Å. Sadabs absorption correction (Tmin 0.819 and Tmax
0.928) and 4217 unique reflections (2θ < 50°, Rint ) 0.0867) were
used in the refinement. Full-matrix least-squares refinement on F 2
converged to RF ) 0.0763 [I > 2σ(I)], 0.1479 (all data) and Rw(F 2) )
0.1785 [I > 2σ(I)], 0.2195 (all data).
Crystal Data of 2b. A light-yellow crystal of approximate dimen-
sions 0.60 × 0.12 × 0.06 mm3 was mounted on a glass capillary. A
total of 6396 reflections was collected with a final resolution of 0.75
Å. Sadabs absorption correction (Tmin 0.783 and Tmax 0.928) and 4529
unique reflections (2θ < 50°, Rint ) 0.0412) were used in the
refinement. Full-matrix least-squares refinement on F 2 converged to
RF ) 0.0583 [I > 2σ(I)], 0.0912 (all data) and Rw(F 2) ) 0.1067 [I >
2σ(I)], 0.1236 (all data).
Crystal Data of 3. Single crystals of 3 were obtained by recrystal-
lization from hexanes/ethyl acetate. A light-yellow crystal of ap-
proximate dimensions 0.40 × 0.20 × 0.05 mm3 was mounted on a
glass capillary. A total of 25 990 reflections was collected with a final
resolution of 0.75 Å. Sadabs absorption correction (Tmin 0.675 and Tmax
0.928) and 7296 unique reflections (2θ < 50°, Rint ) 0.0590) were
used in the refinement. Full-matrix least-squares refinement on F 2
converged to RF ) 0.0611 [I > 2σ(I)], 0.1349 (all data) and Rw(F 2) )
0.1188 [I > 2σ(I)], 0.1500 (all data).
Appendix
Derivation of Eq 2. A competitive equilibrium between
monomer and dimer of oligo-R-aminopyridines (OAP) can be
depicted as 2(OAP) h (OAP)2, and the dimerization constant
Kd for the formation of the OAP dimer can be expressed as
Cp
Kd )
(a)
(C0 - 2Cp)2
where C0 is the initial concentration of OAP, and Cp denotes
the concentration of the OAP dimeric form. The rearrangement
of eq a leads to
C0
Cp
1
)
+ 2
(b)
K C
x
d
p
Under equilibrium the absorbance A at any selected wavelength
can be expressed as
A ) ꢀDCpl + ꢀM(C0 - 2Cp)l ) (ꢀD - 2ꢀM)Cpl + C0ꢀMl
A - C0ꢀMl
Cp )
(c)
(ꢀD - 2ꢀM)l
where l is the cell length, and ꢀM and ꢀD denote the molar
extinction coefficient of monomer and dimer, respectively.
Combining eqs b and c, one obtains
C0l(ꢀD - 2ꢀM)
A - C0ꢀMl
(ꢀD - 2ꢀM)l
)
+ 2
(d)
K (A - C ꢀ l)
x
d
0 M
Dividing (ꢀD - 2ꢀM) on each side leads to eq e.
C0l
l
1
2
)
+
A - C0ꢀMl
ꢀD - 2ꢀM
(e)
K (ꢀ - 2ꢀ )
(A - C ꢀ l)
x
x
d
D
M
0 M
By multiplying ꢀM on each side and applying the Beer-Lambert
law, eq e can be rewritten to obtain eq 2
A0
lꢀ2M
2ꢀM
1
Crystal Data of 4. Single crystals of 4 were obtained by recrystal-
lization from hexanes/ethyl acetate. A light-yellow crystal of ap-
proximate dimensions 0.20 × 0.15 × 0.10 mm3 was mounted on a
glass capillary. A total of 25 402 reflections was collected with a final
resolution of 0.75 Å. Sadabs absorption correction (Tmin 0.796 and Tmax
0.928) and 4346 unique reflections (2θ < 50°, Rint ) 0.0917) were
used in the refinement. Full-matrix least-squares refinement on F 2
converged to RF ) 0.0796 [I > 2σ(I)], 0.1338 (all data) and Rw(F 2) )
0.1410 [I > 2σ(I)], 0.1658 (all data).
)
+
(2)
A - A0
ꢀD - 2ꢀM
K (ꢀ - 2ꢀ )
(A - C ꢀ l)
x
x
d
D
M
0 M
where A0 in eq 2 simply denotes the absorbance of the monomer
assuming that no dimer is formed at the prepared concentration.
ꢀM can be derived by performing the concentration-dependent
study at a sufficiently low concentration so that only monomer
mainly exists. By knowing the proportionality of the dilution,
A0 values can thus be obtained in each prepared concentration.
Knowing ꢀM at the analyzed wavelength, Kd can be deduced
by intercept/(slope)2 ) 2Kd/lꢀM.
Acknowledgment. We thank the National Science Council
and the Ministry of Education of the Republic of China for
financial support.
Supporting Information Available: CIF files containing
detailed crystal data of 2a, 2b, 3, and 4. Original H and 13C
NMR spectra of 2-9 and 10-13. Experimental details for the
NH’s chemical shifts collected from the concentration-dependent
experiments (PDF). These materials are available free of charge
via the Internet at http://pubs.acs.org.
1
(39) SMART V 4.043 Software for the CCD Detector System; Siemens Analytical
Instruments Division: Madison, WI, 1995.
(40) SAINT V 4.045 Software for the CCD Detector System; Siemens Analytical
Instruments Division: Madison, WI, 1995.
(41) Sheldrick, G. M. SHELXL-93, Program for the Refinement of Crystal
Structure; University of Go¨ttingen: Go¨ttingen, Germany, 1993.
(42) SHELXTL 5.03 (PC-Version), Program Library for Structure Solution and
Molecular Graphics; Siemens Analytical Instruments Division: Madison,
WI, 1995.
JA011679P
9
J. AM. CHEM. SOC. VOL. 124, NO. 16, 2002 4297